State of Agent Engineering 2025 by LangChain
- Production momentum is real, with 57% of respondents having agents in production, with large enterprises leading in adoption
- Quality is the production killer, with 32% citing it as a top barrier. Meanwhile, cost concerns dropped from last year
- Observability is table stakes. Nearly 89% of respondents have implemented observability for their agents, outpacing evals adoption at 52%.
- Using multiple models is the norm. OpenAI leads with their GPT models but Gemini, Claude, and open source models see significant adoption. Fine-tuning has not been widely adopted.